Help Me Crowdsource Oz Linden's Explanation of the New Third Party Viewer Policy - What Does It Actually Mean?

Click here for a long .mp3 audio recording of Oz Linden, Director of Open Development at Linden Lab (Scott Lawrence IRL) explaining the changes to the company's policies around third party viewers, which forbid a viewer which alters SL "in such a way that users on other Viewers don't experience the same virtual reality." The policy change has provoked a lot of controversy, in great part because its vague wording. In Oz's presentation, he tries to assure TPV developers that their contributions are valued and wanted, and that the policy change is meant to prevent abuses. At the same time, the developers he engages with seem resentful and distrustful, at least partly due to Linden Lab failings from years that were prior to Oz joining the company. But as it's a long talk, I've only heard snippets so far.

Give it a listen, and if you're a TPV developer or user, post in Comments the parts which give you most concern -- or the most reassurance. Be sure to include time codes!

Hat tip: Ms. Emilly Orr, who has much analysis and commentary here.

Tweet

Comments

[Pussycat Catnap]

Its a little worrying when you need a Guru to understand SL...

If that's OZ's avatar in the image up there... how fitting given current affairs.

[qarl]

if i may be so bold - i think i can easily describe their goal:

they do not want someone to extend SL in a popular way that they themselves cannot control. the vagaries and inconsistencies arise when they try to disguise this goal.

my question is: good lord guys, WHY? you did not have control of the jiggle boobs. you did not have control of parcel windlight settings. you did not have control of multiple attachment points. and yet, despite everything you fear - these situations resolved themselves quite satisfactorily. why fix something that hasn't broken?

...

there are two ways to lead. one is to be awesome in all ways and inspire others to follow you. the other way is to give orders. one of these ways does not work.

[Metacam Oh]

and Mojang which is chugging along just brought in 4 new employees from a popular "third party" mod called Bukkit, and are beginning to implement a modding system.

If that were Linden Lab they would have just shut them down. I hardly see any innovation coming from Linden Lab, and to me if you are as slow and uninspiring as Linden Lab why would you take this approach? What the Lab doesn't do anything new and now they don't want anyone else to do anything new?

Just odd, they should be making it easier for people to innovate and contribute new code instead of the complete opposite. Let us "virtual worlders" shape Second Life, Rodvik you can then turn all your coders to the new text adventure game or whatever.

[TomasG]

LL is just following their motto, as usual.

"IF IT AIN'T BROKE, FIX IT UNTIL IT IS"

[Ignatius Onomatopoeia]

Hamlet, this is a chance to have a bit of fun...we are like pupils trying to unpack a Zen master's koan, and Oz looks the part. So, here goes from a guy paid to do literary exegesis...until I passed out listening to the MP3.

http://iggyo.blogspot.com/2012/02/literary-close-reading-of-oz-lindens.html

[Pathfinder]

http://en.wikipedia.org/wiki/The_Scorpion_and_the_Frog

[Hamlet Au]

"they do not want someone to extend SL in a popular way that they themselves cannot control."

But where in the presentation does he actually say or even imply that?

[Arcadia Codesmith]

I want deformable mesh, and I want it now.

But... I sort of see the point. If I design a deformable mesh outfit that looks great in a TPV, but which looks really pug-ugly in the official viewer, then anybody logging in for the first time on the official viewer and seeing my outfit is going to say, "oh, this is a world with hideously ugly graphical glitches", not "oh, this is a world that allows development freedom to third-party creators even though it doesn't work perfectly with the standard viewer."

I'm a coder myself. I know the frustration keenly. And the Lab really, really needs to clarify what EXACTLY they mean.

But sometimes you've just got to develop to the standard, for the sake of the end user experience, while you patiently push for the standard to be updated.

[qarl]

@Arcadia - similarly - one could create a build which looks like crap without shadows. would this be a good reason to kill early shadow viewers?

[shockwave yareach]

Fortune cookie say: only doctors have infinite patience...

[Pussycat Catnap]

Personally I think this all comes down to still dealing with the remaining lingering impact of the popular malware viewer Emerald. The project moved hand, in theory, and changed names, and got rid of the portion of malware that forced it to change its name...

But it still kept features that invaded privacy. And hoping its "new team" would see to changing this on their own didn't work.

2K also appears aimed at pro-active measures - Emerald teaches us that if given wiggle room, unethical folks will behave unethically. So clamp down hard ahead of time - before say... the Linden Realms game tools go into SL, and provide more tools to be exploited.

Unfortunately, the nature of the impact of policing folks to who fail to self-regulate is that a lot of other innocent victims get caught up in the mess...

[qarl]

@Hamlet - their stated goal is to keep a unified experience. their fear is that if an outsider's feature becomes extremely popular - the default user's experience will become degraded.

but why will it become degraded? because LL themselves have refused to adapt to the popular feature. they have decided not to copy it - they have decided not to improve on it - they have decided not to provide a superior alternative.

they want to maintain the control to say "no, you cannot have this feature even if it is wildly popular. we don't like it and we won't do it."

...

which is fine. that's what many software companies do. but they don't try to sell themselves as an open platform.

[elizabeth (16)]

when discussing this, comparing TPV development to linden development is a bit of an overreach imo. the better comparsion in terms of inworld management of the inworld experience, shared or otherwise, is with LSL developers. LSL developers are fully subject to the managerial dictates of linden. to obtain new features or changes, LSL developers must work entirely through linden processes. seems to me that the general thrust of this policy is to bring the management of TPV development into line with LSL development insofar as it affects the inworld

this may or may not stifle indie TPV development but from a managerial pov of the inworld experience, its the correct position for a company in this position to take imo

[Hamlet Au]

"their fear is that if an outsider's feature becomes extremely popular - the default user's experience will become degraded."

Timecode for when he says this?

[qarl]

> Timecode for when he says this?

that's his whole point - that the "shared experience" will become inconsistent. users without the new features will be at a disadvantage to those with.

[Hamlet Au]

That seems like a reasonable point, doesn't it? If a new user downloads the official, relatively user-friendly client, then finds out that TPV X has Y cool feature, they may then try to use TPV X... even if it's intended for hardcore power users, and is not anywhere near as user-friendly.

End result: a new user who's even more likely to quit. How is that desirable?

[Pussycat Catnap]

Most of the way through the audio at this point.

Its a lot more re-assuring than just the written policy and all the online panic.

Already past the point where they say they want to add in the mesh deformer...

[Missy Restless]

Nice try Hamlet. People! This was a call for a crowdsourced explanation and analysis of Oz's statements with time codes. Seems like everybody is just posting their opinions. Maybe we just have to wait till people have time to listen to it. I loved the exegesis by Ignatius at http://iggyo.blogspot.com/2012/02/literary-close-reading-of-oz-lindens.html

[Ann Otoole InSL]

is the sl client open source anymore? Or is the code no longer compliment and LL simply has a program for voluntary free of charge code contributions?

[Hamlet Au]

"This was a call for a crowdsourced explanation and analysis of Oz's statements with time codes"

Yes, thanks! Quotes! Time codes! Definitive statements now, interpretations later.

[Ezra]

Why not just link to Natales' blog on this? She did an exhaustive good job of adding timestamps to critical parts http://blog.nalates.net/2012/02/26/sl-viewer-policy-change-meeting/

[Hamlet Au]

It'd love to, but it's way too exhaustive for most readers.

Nalates, what are the TOP FIVE most important segments I should highlight, when I link to it?

[Ann Otoole InSL]

Tateru has asked questions and received answers:
http://dwellonit.taterunino.net/2012/02/28/linden-lab-answers-questions-about-the-third-party-viewer-policy-changes/

[Masami Kuramoto]

I welcome the TPV policy changes; they were long overdue. I listened to the podcast, and what Oz Linden said made perfect sense to me. There have been privacy-invading features in third-party viewers, and there has been harrassment based on viewer tags. There have been features in TPVs that made the grid experience unpleasant for everyone else. I and many others pointed this out a long time ago, and finally Linden Lab is doing something about it. Of course the TPV folks are unhappy now. If you want to drain a swamp, don't ask the frogs for an objective assessment of the situation.

Oz mentioned that usage of Viewer 1 is now at 2 percent. So why are 30 percent of the grid's residents still unable to see mesh? It's because TPVs have been slow to adopt that feature in a way that doesn't cripple performance or stability. And yet here we have TPV developers criticising Linden Lab for being slow to adopt groundbreaking TPV features such as bouncing boobs. Oh the irony!

Message to TPV developers: This platform is not yours. The viewer's source code is not yours. You are allowed to inspect and modify it, but on Linden Lab's grid the Lindens are calling the shots. If you don't like that, there's OpenSim for you.

[Ignatius Onomatopoeia]

@Missy, thank you, if only for admitting you did not fall asleep! Ain't hermeneutics some tweedy fun!

I do get it that TPVs are not to be banned. Oz Linden seemed clear enough on that.

Meanwhile, we see where the hammer will come down. I suspect that Viewer 1 has a limited life-span.

[foneco zuzu]

Masami, sorry to say but i have the freedom do choose if i want to see mesh or not!
It's users right to choose waht we wants to see, like reducing graphics to minimum setings and so on.
And just for you to know, on Open sim, i can even decide in my sims, if i allow mesh to be enabled or not!
And if Sl is not ours, i damn be stop paying foe its use as i do and move as i did to Os grid!

[foneco zuzu]

Cause a great rule, that the LAb seems to forget is this 1:
Users are the owners cause they are the ones that pay and keep the business alive!

[Pussycat Catnap]

Frankly going through and timestamping pieces of some audio recording to get an idea of a written policy is near to one of the -worst- ways of reaching an understanding of what is going on.

It can let you make a selective and therefore biased interpretation of what you -think- is Oz Linden's opinion on the policy he's been handed...

But it does nothing to clarify that policy.

Read the policy itself, and consider the terms in it and what they would mean. Anything else is just rumor.

A good example to show why going off the opinions of one random LLer, in an audio no less rather than even something written... is to look back at Blondin Linden and the adult content policy - wherein much of what Blondin stated as policy was in fact incorrect or altered over time to a new interpretation of the written policies.

Timestamping an audio is just a shoddy way of 'dialing it in' when it comes to gaining comprehension of something.

[Noirran Marx]

I consider myself to be an average user of SL, I do a little building, make clothing, and enjoy taking pretty pictures. Why don't I use the LL viewer? Because it crashes every ten minutes. If LL would build a viewer that didn't crash every ten minutes I would use it, until then, I will happily motor along on my stable third party viewer. My experience is just fine thank you very much.

[shockwave yareach]

Noirran: give your computer more ram. You'll be surprised how long the viewer works then. I've run 8+ hours on Viewer3 without even a hiccup. A great many crashes seem to come from the Viewer not cleaning up its ram usage all that well. So if your computer is short on ram to begin with, the poor trash collection makes the computer run out sooner rather than later.

The good part of the story is that putting tons of ram in a computer is very cheap today.

[Adeon Writer]

The policy is vague and really needs to be explained in more absolute terms.

Given Oz's explination of what the policy was meant to say however, I don't have a problem with it. Just go and actually make the policy say that!

[Missy Restless]

One thing I learned listening to this came around 39:00 when Oz Linden points out that Linden Lab's previous TPV policy contained everything in the new policy - these changes were not changes to the policy but an attempt to more explicitly clarify the policy.

One thing I did not understand very well was when, in reply to a question around 25:00 about whether making the display of viewer tags a preference (disabled by default) would violate the policy, Oz Linden replied that it would violate the policy. He repeatedly stated that it would be a violation of the user's privacy to display the viewer tag but how is it a violation of my privacy if I explicitly go into preferences and enable viewer tag display? He avoided addressing this and just stuck to "it's a violation of privacy". He even goes so far as to say that it's ok for me to put my viewer tag in my group tag - so, how is that different than me setting a preference?

Finally, the most interesting and concerning (to me) aspect of the policy is discussed around 49:00. This concerns "shared experience". A lot of what Oz Linden says addressing the TPV developers concern over how this dampens innovation rests on his belief that the Lab and in particular the people at the Lab there now are more receptive and able to effectively collaborate with TPV developers on viewer innovation. Let's hope he is correct in this belief and, if so, it continues to be the case even after he and others move on.

[Masami Kuramoto]

"Masami, sorry to say but i have the freedom do choose if i want to see mesh or not!"

No, you don't. You have the choice to see mesh properly or as a bunch of placeholder prims, but you'll see it either way, no matter which viewer you use.

[shockwave yareach]

Masami: Do you also insist on having to right to not see spheres? How about cubes -- do those offend you somehow too?

[Masami Kuramoto]

I'm not talking about rights but about available options. There is currently no graphical viewer that doesn't show mesh in one way or another, because those that don't support mesh will render placeholder objects instead, whether you like it or not.

But all that is missing the original point anyway. What I was referring to is people complaining about LL being slow to adopt TPV features. They are conveniently ignoring that TPVs have been even slower to adopt mesh.

[shockwave yareach]

The main problem I see with the entire TPV system is that for all that people complain about LL's viewer lacking new features, the moment LL does bring out an improvement, none of the TPV folks wanted to implement it. Just like LL didn't want to implement body physics or multiple attachments. We have a circus of clowns, each trying to prove their shoes are bigger by pretending the other guys are barefoot.

And that's how we end up with new TPV rules from the 2301st floor of the Linden Ivory Tower. What LL should do is say no Release version of a TPV has features different from LL, but Beta versions can demonstrate new stuff on the grid and once proven, those get rolled into everyone's viewers. Considering LL's past history with customers and coders alike, I predict a nice glacial period in Hades before that happens.

[foneco zuzu]

I still say and say it again!
Stop thinking about All Might Lab and start moving to other grids, cause more and more are tired of being treated like dumbs!

[jjccc coronet (@JJcccART)]

Um er was said 1967 times in that meeting by one person if it was cut out of the converation the audio file could have been 37 minutes shorter and even shorter if the gaps were cut out to, but it was kinda interesting to find out that linden labs is about to implement some thing soon, not sure what as the meeting was like listerning to a bunch of butt hurts who dont trust one and other but who cares im just a user not an abuser

[Ayesha Askham]

@Masami

You suggest that the poor percentage of SL that can render Mesh in their viwers is due to the slow adoption of the Mesh-capable render pipeline in V1 type viewers, aka Phoenix et al.

Might it not be that the alternatives are so user UN-friendly and LL's own Viewer 2 developments so resource-hungry that many users migrated to a more user and system friendly V1 some time ago, and the back-porting of the Mesh-capable render-pipeline was a difficult and time-consuming job.

The fact is that LL viewers are still resource-hogs, and the latest evolutions have some serious memory leaks.

Irony? Yes there is irony in Linden Lab's inability to provide a "stable and predictable" experience for its users and its insistence that others abide by its archeological time scales.

[Ayesha Askham]

@Masami

Might not one of the reasons for the small percentage of LLV1 users be because of the appalling resource-hungry mess that was LL's V2?

As a result, many users migrated to more useable TPVs, based on V1.23 or Snowglobe code, and back-porting the Mesh render-pipeline to those codebases took time.

I agree that it took some time for Firestorm to incorporate Mesh, and it is probably fair to say we owe more to Henri Beauchamp that the Firestorm devs for that.
The resource-hog nature of LLs V2 developments and the serious memory leakage issues that the current LL viewer seems to possess both mitigate against uptake by existing users and retention of new ones.
There is indeed irony in the speed at which TPV devs solve issues versus the archeological timescale of Linden Lab change implementation.

[MadBiker Wolf]

Masami Kurmoto,

You are exactly correct when you say "If you want to drain a swamp, don't ask the frogs for an objective assessment of the situation.", but you apply it in the wrong direction. SLis allegedly about openness and creating our own worlds, but Linden is forcing us all to share only their vision, stiffing creativity, and removing freedom of choice.

On the subject of mesh, ALL major third-party viewers now support mesh. The fact that users don't upgrade their viewers isn't the TPV developers fault. You are being disingenuous here.

Secondly, TPV's have provided more than just breast physics. I shan't list them all, but the number of new non-LL features that the user base likes is long and impressive. That's obvious since TPV's make up the majority of viewers being used on the grid.

Lastly, while the platform doesn't belong to anyone but Linden Labs, LL specifically and with intent embraced Open Source. They did this in order to grab some free development. Now that the non-paid developers are out pacing the Linden-paid programmers, Linden Lab is acting like a spoiled child and saying "fine, I'm taking my ball and going home."

If Linden wanted to keep things under their sole control, they shouldn't have embraced Open Source labor.

Linden Labs is not known for making good business decisions and most of these new changes are just another example. They fail to listen to the user base (instead, they have select sycophants who tell them what they want to hear) and thus are losing both corporate and private individuals as paid members who add to the SL economy.

I frankly doubt that there is a single Linden employee who uses SL for more than testing. They don't know what the average person is doing. They don't create things in-world, so they have no clue what builders need or want. They don't try to run an in-world business, so they have no clue what small-business people need.

In short, Linden Labs has very little understanding of what goes on in their virtual world.

(A side note about the alleged TPV bullying that even Jessica Lyons of Phoenix/Firestorm claims to have seen: Lots of us have done the same experiment and the ONLY bullying we saw was people disliking the 1 day old accounts, NOT the viewer being used. That fact that new people almost always use the default Linden viewer just makes it seem that people were picking on the viewer, instead of the annoying 1 day old asking about harassment.)